Reproducibility of Critical Saturation Measurements of Clay

While existing structures present an excellent opportunity for energy savings, it has been established that thermal upgrades to masonry buildings by interior insulation significantly increase the risk of freeze-thaw deterioration of the bricks. Before implementing such retrofits, it is important to understand the risk of damage to the masonry envelope. The critical saturation approach has been proposed as a method for conducting a risk assessment of the freeze-thaw resistance of in-service brick masonry. In order to produce valuable insights, any hygrothermal simulation and analysis relies on the input of accurate material and hygric properties and the outputs must be compared to well established targets. Therefore, it is of interest to verify the reproducibility of the experimental measurements of such properties. Factors such as operator, equipment, and procedural differences can produce deviations in measurements obtained from different laboratories. The critical degree of saturation is the principal metric in this risk assessment framework and is determined by an experimental methodology called frost dilatometry. As the critical degree of saturation is used as the threshold for risk in this limit state design process, it is important to understand the precision of this measured value. Frost dilatometry measurements were conducted at two different laboratories to examine the reproducibility of the critical saturation measurements for 30 brick samples. Dilation measurements are determined with an uncertainty of 50 microstrain. The data analysis methods are examined as a source of discrepancies between lab measurements. Procedural challenges and differences, such as specimen dimensions and number of freeze-thaw cycles, are examined. The saturated moisture content is determined by vacuum saturation. While the results between the two laboratories are generally in agreement, it is a source of error in comparisons of the critical saturation between labs. Correcting for differences in saturated moisture content, it was found that the measurement of the critical degree of saturation are in good agreement. The critical degree of saturation differs by 1-3% when expressed as a ratio of water content to the dry weight of the material indicating the reliability of critical degree of saturation measurements. While differences between the laboratories were small, the uncertainty of the critical degree saturation for both laboratories was large.